ECCO V4r4: Daily Ocean Bolus Transport Streamfunction, 1992-2017

A 26-year daily reconstruction of the Gent-McWilliams ocean bolus transport streamfunction on a high-resolution global grid. This dataset is a dynamically consistent state estimate from NASA's ECCO project, fitting a 1-degree MITgcm model to multiple satellite and in-situ observations. It covers the period from 1992-01-01 to 2018-01-01.

Use Cases

• Validate and improve parameterizations of mesoscale eddy transport in climate models based on the bolus streamfunction.
• Study decadal variability in ocean heat content and carbon storage based on the 26-year reconstructed ocean state.
• Assimilate historical ocean circulation data into new models based on the observationally constrained ECCO framework.
• Analyze the impact of satellite altimetry and other remote sensing data on ocean state estimates.

Strengths

• Integrates over 25 years of diverse observational data, including satellite altimetry, sea surface temperature, salinity, and in-situ profiles.
• Provides a dynamically and kinematically consistent 3D reconstruction of the global ocean and sea-ice state.
• Covers a long temporal range from 1992 to 2018 on a high-resolution native LLC90 model grid.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Data is provided in multiple non-standard file formats (JPEG, BIN, ISO, HTML, PDF), which may complicate direct analysis.
• Row count and dataset size are unknown, which may limit suitability assessment for large-scale ML projects.

Provenance

Source

National Aeronautics and Space Administration (NASA) via the Estimating the Circulation and Climate of the Ocean (ECCO) project.

Collection Method

Least-squares fit of a free-running 1-degree MIT general circulation model (MITgcm) to a wide array of satellite and in-situ observations.

Time Range

1992-01-01 to 2018-01-01

Freshness

Last updated 2026-03-13 03:05:48.256879; freshness should be verified.

Geography

Global ocean coverage on the LLC90 grid.